|
Download README.md from davidzhou302/ActiveScale: direct link, hf CLI and curl.
- Browser
- Download file 2.68 kB
-
https://huggingface.co/davidzhou302/ActiveScale/resolve/main/README.md
- Command line
-
hf download hf://davidzhou302/ActiveScale/README.md
-
curl -L -o README.md https://huggingface.co/davidzhou302/ActiveScale/resolve/main/README.md
2.68 kB
| library_name: openpi | |
| license: other | |
| license_name: gemma-terms | |
| license_link: https://ai.google.dev/gemma/terms | |
| pipeline_tag: robotics | |
| <h1 align="center">ActiveScale: Scaling Active Perception for Robots<br>across Model, Data, and Hardware</h1> | |
| <p align="center"> | |
| <a href="https://arxiv.org/abs/2609.18514"><img src="https://img.shields.io/badge/arXiv-B31B1B?style=for-the-badge&logo=arxiv&logoColor=white" alt="arXiv"></a> | |
| <a href="https://active-scale.github.io/"><img src="https://img.shields.io/badge/Project_Page-4285F4?style=for-the-badge&logo=googlechrome&logoColor=white" alt="Project page"></a> | |
| <a href="https://youtu.be/Iya0bSZ8Dko"><img src="https://img.shields.io/badge/YouTube-FF0000?style=for-the-badge&logo=youtube&logoColor=white" alt="YouTube"></a> | |
| <a href="https://huggingface.co/davidzhou302/ActiveScale"><img src="https://img.shields.io/badge/Hugging_Face-FFD21E?style=for-the-badge&logo=huggingface&logoColor=black" alt="Hugging Face"></a> | |
| <a href="https://modelscope.cn/datasets/shuai302/ActiveScale"><img src="https://img.shields.io/badge/ModelScope-624AFF?style=for-the-badge" alt="ModelScope"></a> | |
| </p> | |
| <p align="center"> | |
| <a href="https://youtu.be/Iya0bSZ8Dko"> | |
| <img src="https://img.youtube.com/vi/Iya0bSZ8Dko/maxresdefault.jpg" alt="ActiveScale video" width="100%"> | |
| </a> | |
| </p> | |
| <p align="center"> | |
| <a href="https://active-scale.github.io/"> | |
| <img src="https://raw.githubusercontent.com/active-scale/active-scale.github.io/main/assets/teaser_1.webp" alt="ActiveScale overview" width="100%"> | |
| </a> | |
| </p> | |
| <p align="center"> | |
| ActiveScale presents a scalable way to study active perception for robot manipulation across model, data, and hardware.<br> | |
| It equips a vision-language-action model with temporal camera observations and camera-pose prediction,<br> | |
| and introduces the Active-perception Mobile-manipulation Platform (AMP) for coordinated viewpoint and manipulation control. | |
| </p> | |
| <h2 align="center">Checkpoints</h2> | |
| <table align="center"> | |
| <thead> | |
| <tr><th>Checkpoint</th><th>Description</th></tr> | |
| </thead> | |
| <tbody> | |
| <tr><td><code>midtraining</code></td><td>Mid-training checkpoint</td></tr> | |
| <tr><td><code>bag</code></td><td>Bag task checkpoint</td></tr> | |
| <tr><td><code>drawer</code></td><td>Drawer task checkpoint</td></tr> | |
| <tr><td><code>pot</code></td><td>Pot task checkpoint</td></tr> | |
| <tr><td><code>box</code></td><td>Box task checkpoint</td></tr> | |
| <tr><td><code>under</code></td><td>Under-table task checkpoint</td></tr> | |
| </tbody> | |
| </table> | |
| <p align="center"> | |
| Code and usage: <a href="https://github.com/ShuaiZhou302/ActiveScale">github.com/ShuaiZhou302/ActiveScale</a> | |
| </p> | |